Re: [BUG] org-copy-subtree in a file with local variables marks buffer as modified [9.5.3 (release_9.5.3-6-gef41f3 @ /home/ignacio/repos/emacs/lisp/org/)]

[Ihor Radchenko <yantar92@gmail.com>]

Ignacio Casso <ignaciocasso@hotmail.com> writes:

> What would you think about a change like the following in the definition of
> `org-preserve-local-variables' to solve the issue?
>
> + (let ((modified-before-p) (buffer-modified-p))
>     ;; current code that deletes local variables
> +   (unless modified-before-p (set-buffer-modified-p nil))
>     ;; current code that executes body
> +   (let ((modified-after-p (buffer-modified-p)))
>       ;; current code that restores local variables
> +     (unless modified-after-p (set-buffer-modified-p nil))))
>
> Could the current code or my proposed change have any more unintended
> consequences, for example in the undo tree or mark ring? Sometimes when
> I undo or jump to the mark, the point moves to the last visible, usually
> folded headline of the file, even if I had not edited it recently, and I
> have always assumed that it's because it wants to go to the end of the
> buffer for some reason I've never been able to come up with or
> debug. Could that have something to do with this?

buffer-modified-p is not the only parameter affected by juggling around
the local variables. There will be undo history,
buffer-chars-modified-tick, before/after-change-hooks triggered by
`org-preserve-local-variables', etc.

However, I do not see any obvious way how your proposed change can
negatively affect all the above. Feel free to propose a patch.

Best,
Ihor